osmander's profile - activity

Hello

I am trying to use stereo matching implementations of OpenCV. (StereoBM,StereoSGBM). But i am having some problems. The resulting disparity maps are always crop from one side.

OpenCV assumes that the disparity between a pair of rectified stereo image is always to left (is this correct?). I have some pairs where the disparity is both to the right and to the left. For the upper part of images, it is to the right then around middle it becomes almost zero and it switches towards left.

Therefore i have to assign a large disparity interval. This results in crop disparity images. Do you have any idea how to apply StereoBM or StereoSGBM for images like the ones i have?

Thanks

Thanks for the answer. Do you think it is a good idea to map the pixel values of an CV_16U image to 0-255 and then apply SURF detector?

Hello all

I have some images with gray values higher than 255 so i keep them in CV_16U type cv::mats. But i realised that surf descriptors need CV_8U images as input. Is this correct?

When i convert them to CV_8U, obviously i am losing some data. So is there a way to get the SURF points for non CV_8U images?

Thanks

Thanks for your answers. identify command will be very handy!

I guess, to read a part of an image without opening the whole image, i should write my own reader class.

Cheers

Hello everyone

i am dealing with very big images(>2gb). i am having some memory problems.

so i wonder if there is way to get the image height without loading the images? now i am using cv::imread() and get the width by mat.rows, which is not very convenient for me.

and my second question would be if there is way to open only a portion of images? for this purpose i open the images again with cv::imread() and get the parts with cv::Mat(imageData, lineRange, sampleRange). i want to directly open the specific parts which are define by lineRange and sampleRange( whose are cv::Range).

Thanks